Character Recognition Tamil Language in Printed Images using Convolutional Neural Network (CNN) analysis

In this paper, we suggested a system for handwritten character recognition in printed images of the Tamil language. The current work is being implemented using Optical character Recognition (OCR) in step one of the projects. The most recognized issues are poor print and paper quality and unknown font faces. OCR is also not accurate in acknowledging the handwritten text and the fonts. Also, the implementation is carried out using the Convolutional Neural Network (CNN) model with handwritten digit recognition. CNN has the potential to recognize handwritten picture characters clearly and robustly. For Tamil handwritten character classification, we have considered the CNN in this paper without any feature collection. In terms of test accuracy, the proposed approach provides comparable output with the other existing methods. And it was checked on a major data set as well. For Tamil handwritten character recognition, experiments on a large data set showed the robustness of this model. The outcome of the proposed model for handwritten Tamil character recognition using CNN gives an accuracy of 98.00%